List your company

Mathematical Reasoning on AIME 24 (Accuracy, Pass@8)

70AccuracyQwen3-8B (Thinking)
-2.38416.40835.253.992May 16, 2026

Evaluation results

MethodLinks
Qwen3-8B (Thinking)Model=Qwen3-8B, Variant=Thinking2026.057083.3
Qwen3-1.7B + Guided (ρ¯ ≈ 0.16)Model=Qwen3-1.7B, Intervention Rate (rho)=≈ 0.162026.054573.3
Qwen3-1.7B (Thinking)Model=Qwen3-1.7B, Variant=Thinking2026.0539.273.3
Qwen3-0.6B + Guided (ρ¯ ≈ 0.13)Model=Qwen3-0.6B, Intervention Rate (rho)=≈ 0.132026.0532.570
Qwen3-1.7B + Guided (ρ¯ ≈ 0.06)Model=Qwen3-1.7B, Intervention Rate (rho)=≈ 0.062026.0520.446.7
Qwen3-0.6B + Guided (ρ¯ ≈ 0.04)Model=Qwen3-0.6B, Intervention Rate (rho)=≈ 0.042026.051036.7
Qwen3-0.6B (Thinking)Model=Qwen3-0.6B, Variant=Thinking2026.059.230
Qwen3-1.7B-BaseModel=Qwen3-1.7B, Variant=Base2026.05520
Qwen3-0.6B-BaseModel=Qwen3-0.6B, Variant=Base2026.050.43.3